Abstract
Enhancers function as DNA logic gates and may control specialized functions of billions of neurons. Here we show a tailored program of noncoding genome elements active in situ in physiologically distinct dopamine neurons of the human brain. We found 71,022 transcribed noncoding elements, many of which were consistent with active enhancers and with regulatory mechanisms in zebrafish and mouse brains. Genetic variants associated with schizophrenia, addiction, and Parkinson’s disease were enriched in these elements. Expression quantitative trait locus analysis revealed that Parkinson’s disease-associated variants on chromosome 17q21 cis-regulate the expression of an enhancer RNA in dopamine neurons. This study shows that enhancers in dopamine neurons link genetic variation to neuropsychiatric traits.
Main
To date the majority of disease- and trait-associated variants emerging from genome-wide association studies (GWAS) of neurologic and psychiatric diseases lie within nonprotein coding sequences. Several lines of evidence suggest that a proportion of such variants are involved in transcriptional regulatory mechanisms, including modulation of enhancer elements1. Many regulatory elements are cell-type-preferential2,3, and therefore sequence variants with functional consequences are expected to manifest their effects more strongly in the cell type(s) most relevant to a specific disease phenotype.
Here we focused on systematically identifying all noncoding regulatory elements transcriptionally active in a morphologically, functionally, and biochemically distinct neuronal archetype: dopamine neurons of the substantia nigra pars compacta in human midbrain. We hypothesized that genetic variation associated with diseases involving dopaminergic neurotransmission exerts its effects through modulation of enhancers functionally active in this particular type of neurons. Perturbations of the dopaminergic system are important in the pathogenesis and treatment responses of many increasingly prevalent complex genetic diseases, including Parkinson’s disease (PD), which affects 0.5 million people4, schizophrenia, which affects 2.2 million people5, and addiction, which affects 23.5 million people6 (all numbers are for the United States alone). In healthy people, these dopaminergic neurons shape how we conduct our everyday lives, encoding activities related to motivation and reward. Signals from these neurons to the striatum have a profound impact on action learning and automatic movements, while projections to hippocampus and prefrontal cortex influence memories and behavior7.
Our analysis is powered by an integrated hardware–software solution for comprehensively detecting noncoding transcription in one single and minuscule RNA sample and mapping the variation in noncoding transcription to genetic variation within dopamine neurons across multiple individuals. This method combines the base-pair resolution and a comprehensive genome-wide view afforded by ultradeep, total RNA-sequencing with the positional and cytoarchitectural precision afforded by traditional light microscopy.